Morphosyntax of Modals and Quasimodals(1) - DSpace at Waseda ...

Morphosyntax of Modals and
Quasimodals
Introduction
Ryo OTOGURO
Modality is a grammatical category where language’s morphology,
syntax, semantics and pragmatics meet. Due to its ubiquitous nature,
linguistic phenomena covered by the term ‘modality’ are diverse.
Traditionally modality is often defined as the expression of the
speaker’s attitude, opinions or emotions towards the proposition, or
more generally the subjective expression e.g. Lyons , Bybee et al.
. For instance, a pair of sentences in are concerned with the
speaker’s judgement of the proposition that Kate is at home. The pair
in , on the other hand, are relevant to the speaker’s attitude
towards a potential future event, namely Kate’s coming in. The former
is called epistemic modality, whereas the latter is called deontic
modality Palmer : .
a. Kate may be home now.
It is possible possibly the case that Kate is at home
now.
Morphosyntax of Modals and Quasimodals +.*
b. Kate must be home now.
It is necessarily the case that Kate is at home now.

+-3
a. Kate may come in now.
It is possible for Kate to come in now.
b. Kate must come in now.
It is necessary for Kate to come in now.
Although epistemic and deontic modality is encoded explicitly by
modal auxiliaries in and , the definition based on
subjectivity do not stop us from including a wider variety of syntactic
constructions as modal expressions than expected. As Narrog
points out, passive voice and certain aspectual expressions may
convey the speaker’s attitude towards the propositions Kroeger :
, Narrog :, .
Morphosyntax of Modals and Quasimodals
a. Korede amenidemo huraretara,
thison rainDATEXPL rainPASSCOND
sanzandesuyo.
aufulCOP.NPSTEXCL
‘If we got rained on, that would be awful.’
b. Tomo kematian anak.
Tomo die.ADV child
‘Tomo suffered the death of a child.’
You’re telling me you don’t love me anymore.
a and b are adversative passives in Japanese and Malay
respectively. Unlike ordinary passive formation, which involves

decrease of the valency of the predicate, an adversative passive
pragmatically adds the speaker’s negative feeling to the literal
meaning of the proposition. is what Wright calls
experiential progressives. The progressive aspect may convey the
speaker’s epistemic stance with regard to the interlocutor’s utterance.
In , the speaker negatively evaluates the interlocutor’s
utterance of not loving the speaker anymore. Those constructions
clearly deliver the speaker’s subjective evaluation of the proposition,
and according to the definition of modality given above, they may be
categorised as modal expressions.
The fact that a language has various means to express the speaker’s
subjectivity, however, does not mean all of them are directly encoded
in the language’s grammatical system. As argued in a number of
studies, linguistically more salient definition of modality is offered
based on ‘factuality’, namely realis/irrealis distinction e.g. Lyons
, Kiefer , Mithun , Palmer , Narrog . Oneof the
crucial aspects of factualitybased definition of modality is its
independence from the speaker’s point of view since factuality is a
semanticoriented notion. Narrog :, for instance, proposes
the following definition :
Morphosyntax of Modals and Quasimodals +-2
Modality is a linguistic category referring to the factual
status of a state of affairs. The expression of a state of affairs
is modalised if it is marked for being undetermined with
respect to its factual status, i.e. is neither positively nor
negatively factual.

+-1
This definition properly excludes the cases like and as both
adversative passives and experiential progressives are clearly factual,
which is outside the domain of modality.
The exponents encoding indeterminacy of factuality still varies
across languages. In English examples and , for example, the
exponents are modal auxiliaries. Germanic languages dominantly use
auxiliary verbs to express modality, such as English can, German
können, Norwegian kan and Frisian kin, all of which are historically
derived from Gothic kunnan ‘know’ Harbert . Another pattern is
verb inflection, often called mood, as in indicative and subjunctive
forms in Spanish and an irrealis form in Amele
Palmer : .
Morphosyntax of Modals and Quasimodals
a. Creo que aprende.
I believe that learn..SG.PRES.IND
‘I believe that he is learning.’
b. Dudo que aprenda.
I doubt that learn..SG.PRES.SUBJUNCT
‘I doubt that he’s learning.’
Ho bubusaleb age qoqagan.
pig SIMrun out.SG.DS.IRR .PL hit.PLFUT
‘They will kill the pig as it runs out.’
Although the surface descriptive fact is relatively straightforward,
that is auxiliaries are syntactically independent lexical items that take

Morphosyntax of Modals and Quasimodals +-0
a verbal complement whereas verb inflectional formatives are
realisations of morphological categories constructing paradigmatic
organisation of a given lexeme, it is challenging for any grammatical
theories to capture the morphosyntactic nature of such a diverse
range of encoding patterns. With regard to auxiliaries, they are
obviously syntactically independent units, but whether they are
genuine argumenttaking predicates or not requires a careful
inspection of the empirical data. Another issue is phrase structure
wordhood of the formatives. Although some verbal inflectional forms
are clearly affixes as illustrated in Spanish and Amele examples, there
are marginal cases in which the boundary between auxiliarylike
elements and affix/cliticlike elements are not straightforwardly
identifiable. The primary aim of the present study is to examine
whether a unified framework for modality can be provided in
constraintbased parallel grammatical architecture, Lexical Functional
Grammar Bresnan . To this end, I will briefly overview the
historical background of the analyses that have been proposed in the
generative literature in section . Ithen present the insightful data in
Japanese, focusing on the categorical differences between socalled
genuine modals and quasimodals in section . In section , I shall
summarise the previous LFG account for English modals, and then
illustrate how LFG gives us a unified framework for Japanese data as
well. The discussion will be concluded in section .
Feature or predicate?
Modals along with tense and aspect have attracted much attention
since early days of generative syntax. As summarised in Falk ,

+-/
the analyses are mainly divided into two types : AUX feature analysis
and AUX predicate analysis. The AUX feature analysis is originally
proposed by Chomsky and its fundamental assumption is that
an auxiliary is an element that merely contributes tense, aspect and
mood features to the predicateargument relationships specified by
the lexical verb. Therefore, a cluster of auxiliaries are located under a
single node called AUX as in .
Morphosyntax of Modals and Quasimodals
is a phrase structure tree for a sentence, the author may have
written that novel. The tense, modal and aspectual auxiliaries alongside
averbal suffix are located all under the AUX node.
According to the development of transformational frameworks, the
AUX feature analysis has been revised, in which an AUX element is
regarded as a head of the entire clause, namely the Inflection heads
the IP projection S. IP is called functional projection and later
further divided into a layer of functional projections, AgrS, AgrO and
T Pollock . Some recent version of transformational grammar,
called Cartography of grammar, proposes even more fine grained
functional projections. Cinque , for example, assumes a
structure in which various types of Modal, Tense and Aspect

are organised in a hierarchical way ; a simplified version can be
represented as in the subscripts represent different types of
functional projections.
Despite the different structural organisations, namely a single AUX
node and a layer of functional projections, the essential part of the
AUX feature analysis is the claim that features like tense, aspect, and
modality are all functional, i.e. nonpredicative, elements in the
syntactic structure.
The AUX predicate analysis, on the other hand, assumes that an
auxiliary is one type of predicates and takes a complement as in
cf. Ross
Morphosyntax of Modals and Quasimodals +-.

+--
Morphosyntax of Modals and Quasimodals
In contrast to , the two auxiliaries both take a sentential
complement. The subject, the author, originates in a lower subject
position of the lexical verb, written, moves to the subject position of
the lower auxiliary, have, and further moves up to the subject position
of the higher auxiliary, may. Hence, the AUX predicate analysis
regards auxiliaries a kind of raising predicates.
Although the recent transformational grammar dominantly follows
the AUX feature analysis as most dramatically exemplified in , in
principle the choice between the two competing analyses is
determined based on conceptual arguments and empirical observation,
and it often varies across languages and even within a single
language. For instance, Falk argues that an English copulative
progressive auxiliary must be regarded as a predicate, citing
Jackendoff’s , illustration of paradigmatic contrast of a
progressive auxiliary to other lexical verbs as in and the
availability of a PP complement both for the lexical verbs and the
progressive auxiliary in a parallel fashion as in .

a. The children were eating breakfast.
being in the state of eating breakfast
b. The children started eating breakfast.
entering the state of eating breakfast
c. The children kept eating breakfast.
continuing the state of eating breakfast
d. The children stopped eating breakfast.
leaving the state of eating breakfast
a. John kept Bill running/at a run.
b. Moe went on working/with his work.
c. Rodgers is working/at work on a new play.
The empirical data in and seem to suggest that the English
progressive auxiliary is able to take a complement and this fact
cannot be captured by the AUX feature analysis.
Norwegian provides even more robust evidence for the predicative
nature of auxiliaries. As illustrated in , apronominal can appear
as a complement of a modal and tense auxiliary in Norwegian. The
data strongly suggest that modal and tense auxiliaries in Norwegian
are fullfledged argumenttaking predicates. Dyvik , cited in
Falk .
Morphosyntax of Modals and Quasimodals +-,
a. Jeg vi/kan/ma˚/skal dette.
I will/can/may/shall this
‘I want/am able to do/am obliged to do/have a duty to do

+-+
this.’
b. Vil det regne? Det vil det.
will it rain? it will that
‘Will it raining? It will that.’
Romance aspectual auxiliaries, on the contrary, do not behave like
predicates. For instance, although Spanish allows an infinitival
complement to be fronted leaving the main verb behind as in a,
the same operation is not available for the complement of a perfective
auxiliary as illustrated by the ungrammaticality of b Schwarze
, cited in Falk .
Morphosyntax of Modals and Quasimodals
a. Ver el castillo, lo quiere.
to see the castle it.ACC want..SG
‘To see the castle, he wants it.’
b. Visto el castillo, lo ha.
seen the castle it.ACC AUX..SG
‘Seen the castle, he has it.’
The empirical observation reveals that whether auxiliary elements
should be analysed as genuine argumenttaking predicates or mere
feature contributors is largely an itemspecific matter. This suggests
that grammatical theory must be able to provide a framework that
accommodates both types of analyses in its design.
Modals and quasimodals
As pointed out in section , the distinction between auxiliaries and

verbal affixes is not always clearcut. Japanese is one of the
languages that develop a diverse range of modal expressions, and on
the surface, they look all similar Nitta : , Ueda :
.
a. amewa furumai.
rainTOPIC fallNAG.CONJECT
‘It may not rain.’
b. asuwa harerudesyoo.
tomorrowTOPIC sunnyCONJECT.POL
‘Probably, it will be sunny tomorrow.’
c. hayaku tabero/tabetamae/tabenasai.
quick eatIMP
‘Eat quickly.’
Morphosyntax of Modals and Quasimodals +-*
d. watasiga kono hono yakusoo.
INOM this bookACC translateINTENT
‘I am willing to translate this book.’
a. Tarooga daigakue singaku sisooda.
TaroNOM universityto enter doREPORT
‘It seems to be the case that Taro will go to the university.’
b. Taroowa daigakue singaku subekida.
TaroTOPIC universityto enter doOBLIG
‘Taro should go to the university.’
c. syatyooga kuruyooda.
CEONOM comeCONJECT
‘The CEO will probably come.’

+,3
The italicised exponents all encode modality and follow a lexical
verb. However, as Nitta points out, the modal elements in
and are called ‘genuine modals’ and ‘quasimodals’
respectively ,and they exhibit different morphosyntactic behaviours
as summarised in table .
Modal Quasimodal
Tense distinction No Yes
Polarity distinction No Yes
Stacking No Yes
Firstly, quasimodals can either be past or nonpast tense forms while
genuine modals cannot. illustrates that quasimodals have tense
distinctions, sooda/soodatta and bekida/bekidatta.
a. Taroowa daigakue singakusisooda/
TaroTOPIC universityto enter doREPORT.NONPAST/
soodatta.
REPORT.PAST
‘It seems/seemed to be the case that Taro will/would go to
the university.’
b. Taroowa daigakue singaku subekida/
TaroTOPIC universityto enter doOBLIG.NONPAST/
bekidatta.
OBLIG.PAST
Morphosyntax of Modals and Quasimodals
Table . Modal and quasimodal
‘Taro should go/should have gone to the university.’

Secondly, quasimodals can be negated as shown in , whereas
the negative polarity can only be expressed by inherently negative
formatives in genuine modals cf. a.
a. Taroowa daigakue singaku sisoodenai.
TaroTOPIC universityto enter doREPORT.NEG
‘It doesn’t seem to be the case that Taro will go to the
university.’
b. Taroowa daigakue singaku subekidenai.
TaroTOPIC universityto enter doOBLIG.NEG
‘Taro should not go to the university.’
Finally, quasimodals can appear multiple times as long as the
semantic combinations are interpretable as in a. However, such
stacking of modals is impossible for genuine modals as shown in .
a. syatyooga koozyooni kurutumoridattarasii.
CEONOM factoryto comeINTENT.PASTREPORT
‘I heard that the CEO was willing to come to the factory.’
b. kono yamani noboroomai.
this mountainLOC climbINTENTNEG.CONJUNCT
‘It may not be the case that we are willing to climb this
mountain.’
Morphosyntax of Modals and Quasimodals +,2
All those morphosyntactic behaviours indicate that quasimodals
are syntactically independent predicative verbs in that they inflect for
tense and polarity in the same way as other ordinary lexical verbs. In

+,1
addition, the stacking fact strongly supports the idea that they take a
syntactic complement as assumed in the AUX predicate analysis.
The genuine modals, on the other hand, exhibit the properties of
morphological products. The past/nonpast tense distinction
expresses declarative mood, which is in the same morphosyntactic
category, MOOD, asother genuine modals. It also explains why genuine
modals cannot cooccur with each other. Since they participate in the
same inflectional category, the cooccurrence of more than one
formatives causes the clash of the features in that category. In
addition, negative polarity is dominantly realised by a derived
negative lexeme, so that it no longer participates in the verb’s
primary inflectional paradigm Otoguro , .
Lexical Functional analysis
The preceding sections have revealed that modality is expressed in
diverse ways both in terms of their formal encoding, i.e. auxiliaries vs
verbal affixes, and their contribution to the syntactic structure, i.e.
features vs predicates. To account for the behaviours modal
expressions show, I will present an analysis in the grammatical
framework of Lexical Functional Grammar LFG, particularly
focusing on the contrast between genuine modals and quasimodals in
Japanese.
Morphosyntax of Modals and Quasimodals
LFG is a constraintbased unification grammar in which different
linguistic structures are postulated in a parallel fashion. The surface
organisations of lexical items are represented in constituentstructure
cstructure where words and phrases are grouped in a phrase
structure tree and encode linear precedence and hierarchical

dominance relationships. Alongside cstructure, grammatical relations
GF :grammatical function in LFG and other syntactic features are
described in another structure called functionalstructure f
structure. Fstructure is represented as an attributevalue matrix
AVM, inwhich a pair of attribute and value for a given feature is
listed. For instance, a sentence, the children may eat breakfast, is
represented in cstructure as in a and the corresponding f
structure is given as in b.
Morphosyntax of Modals and Quasimodals +,0
The attribute and value pairs represented in fstructure come from
the lexical items located in the terminal nodes in cstructure. For
example, is sample lexical entries of the items. Each entry
consists of a bundle of features called functional description as well as
a phrase structure categories such as Infl, V, D and N. is an
abbreviation of function fM in which f maps one cstructure
node to fstructure, M maps one cstructure node to its mother node,
and refers to the ‘current node’ in cstructure as defined in .
That is, TENSE FUTURE means that in the fstructure
corresponding to the mother node of will, i.e. an I node, the value of

+,/
TENSE attribute is FUTURE.
a. will I TENSE FUTURE
b. the D DEF
c. children N PRED ‘children’
d. eat V PRED ‘eatSUBJ,OBJ’
: f M
:f
The mathematical functions, and , are also annotated on the c
structure, so that the attribute and value pairs associated with lexical
items are correctly mapped onto fstructure. The annotated c
structure and the corresponding fstructure are illustrated as in .
Morphosyntax of Modals and Quasimodals
One of the crucial aspects of the analysis in is the treatment of
the tense auxiliary, will. As specified in the annotations on the I node

and the VP node, i.e. ,the auxiliary and the VP is mapped onto
the same outermost fstructure. As a result, the auxiliary merely
contribute the TENSE feature to the corresponding fstructure while it
takes a VP complement in the cstructure. Hence, the separation of
surface phrase structure and grammatical feature representation, LFG
allows us to state the fact that the auxiliary heads the phrase
structure constituent by taking a VP complement while functionally
it simply adds tense information to the sentence.
Following the observation made in section , Falk proposes
the LFG analysis of the progressive auxiliary along the line of
and .
Morphosyntax of Modals and Quasimodals +,.
a. were I PRED ‘beXCOMPSUBJ’
TENSE PAST
SUBJ XCOMP SUBJ
b. eating V PRED ‘eatSUBJ,OBJ’
ASP PROG

+,-
Morphosyntax of Modals and Quasimodals
Unlike the future tense auxiliary will, were has its own PRED feature,
which suggests that it is functionally an argumenttaking predicate.
Furthermore, one of its governable GFs is an open complement called
XCOMP. An XCOMP lacks one of its GFs required by the local predicate,
eating in the case of . Such a missing GF is identified with one of
the GFs in the outer fstructure. This is attained by the mechanism
called functional equation, SUBJ XCOMP SUBJ, inthe lexical
entry of were. This equation ensures that the value of SUBJ in the f
structure corresponding to the I is the same as the value of the
XCOMP’s SUBJ. This identification of the values of two GFs is represented
by the line connecting them in fstructure.
Another notable feature of this analysis is a nonthematic SUBJ

argument of the PRED value of were in a. The reason that SUBJ is
outside the angled brackets is that it is semantically vacuous to the
predicate. Hence, an XCOMP is the sole semantic argument and the
main role of the SUBJ is to function as a thematic argument of the
inner lexical verb.
The essence of this analysis is that the cstructure is minimally
different from , so that it also captures the fact that the
progressive auxiliary were takes a VP complement in the same way as
will. Still, the corresponding fstructure significantly differs from the
AUX feature analysis.
I will now present an analysis of Japanese modals and quasi
modals. An example of genuine modals is given as in and a
lexical entry of the modal expression is specified as in . As
illustrated in section , the genuine modals in Japanese are inflectional
suffixes, so the entries are given based on the fully inflected forms,
that is the suffixes contribute the value of MOOD to the verbs’ lexical
entries.
Morphosyntax of Modals and Quasimodals +,,
watasiga kono hono yakusoo.
INOM this bookACC translateINTENT
‘I am willing to translate this book.’
yakusoo V PRED ‘yakusSUBJ,OBJ’
MOOD INTENT
The cstructure and the corresponding fstructure can be represented
as in . Since the genuine modal, oo, does not have an

+,+
independent cstructure node, its feature is mapped onto fstructure
together with the lexical verb’s PRED.
Morphosyntax of Modals and Quasimodals
We now look at quasimodals. Since the quasimodals are
independent lexemes inflecting for tense and polarity and taking a
complement, they have a PRED feature subcategorises for a thematic
open complement, XCOMP, and a nonthematic SUBJ. Similarly to the
raising verbs, it requires an equation SUBJ XCOMP SUBJ,
which identifies the nonthematic SUBJ with the SUBJ of the inner
PRED. Therefore, the lexical entries for the two quasimodals
appearing in can be given as in .
sytyooga susio taberu tumoridatta rasii.
CEONOM sushiACC eat INTENET.PAST REPORT
‘I heard that the CEO was willing to eat sushi.’
a. rasii Aux PRED ‘rasiXCOMPSUBJ’
TENSE NONPAST
SUBJ XCOMP SUBJ

. tumoridatta Aux PRED ‘tumoridXCOMPSUBJ’
TENSE PAST
SUBJ XCOMP SUBJ
Those lexical entries allow us to construct the cstructure and the
corresponding fstructure for the sentence as in .
Morphosyntax of Modals and Quasimodals +,*
Iassume that quasimodals are adjoined to VP in cstructure, and the
complement VP is mapped onto the value of XCOMP in the fstructure.

++3
The resultant fstructure contains multiple layers of XCOMPs, whose
nonthematic SUBJs are identified with the innermost SUBJ. Note that
each layer has a distinct TENSE feature, since the quasimodals
independently inflect for tense feature.
One intriguing consequence the present analysis predicts is the
relevance between an XCOMP and the demonstrative, sono. As pointed
out by Inoue , averbal proform soo is used in a reply referring
to the quasimodal in a question. For instance, in the following
conversation, soodesu refers to the quasimodal, tumoridesu as shown
in . Toconfirm this observation, if we use soodesu as a reply to
the question without a quasimodal, it will be unacceptable as in
B . Tomake the reply acceptable, the speaker needs to repeat the
main verb as in B . Even when a genuine modal is used in the
question as in A, soodesu cannot be used as a reply.
A: kimiwa asu daigakuni kuru
youTOPIC tomorrow universityto come
tumoridesuka?
INTENT.POLQ
‘Are you thinking about coming to the university
tomorrow?’
B: hai, soodesu.
yes
Morphosyntax of Modals and Quasimodals
‘Yes, I am thinking about doing so.’
A: kimiwa kesano simbuno
youTOPIC this morningGEN newspaperACC

yomimasitaka?
read.POL.PASTQ
‘Did you read this morning’s newspaper?’
B :??hai, soodesu.
Yes
B :hai, yomimasita.
yes, read.POL.PAST
‘Yes. I read it.’
A: Taroowa ikudesyooka?
TaroTOPIC goCONJECT.POLQ
‘Is Taro going there ?’
B :??hai, soodesu.
yes
B :hai, ikudesyoo.
yes, goCONJECT.POL
‘Yes, he is going.’
This empirical fact indicates that a verbal proform soo corresponds to
the fstructure whose PRED is originally a quasimodal. In addition,
the fact that soodesu cannot refer to the modality expressed by a
genuine modal supports the claim that genuine modals in Japanese
are not independent lexical items.
Morphosyntax of Modals and Quasimodals ++2
A further intriguing aspect is found in the use of a genitive
pronoun sono. Acomplement originally introduced by a lexical verb
can be pronominalised by sono and used as a nominal complement of
a quasimodal. This point can be illustrated by the following

++1
examples.
A: kimiwa asu daigakuni kuru
youTOPIC tomorrow universityto come
tumoridesuka?
INTENT.POLQ
‘Are you thinking about coming to the university
tomorrow?’
Morphosyntax of Modals and Quasimodals
B: hai, sono tumoridesu
yes it INTENT.POL
‘Yes, I am thinking about doing so.’
A: Taroowa kesa tyoosyokuo tabemasitaka?
TaroTOPIC this morning breakfastACC eat.POLQ
‘Did Taro have breakfast this morning?’
B: hai, sono hazudesu.
yes it CONJECT.POL
‘Yes, he must have had that.’
In B, asu daigakuni kuru ‘coming to the university tomorrow’ is
pronominalsed by sono and becomes a complement of the quasi
modal, tumoridesu. Even when a quasimodal is not introduced in the
question, it can be introduced in the reply by taking a pronominal
sono as a complement as shown in B where the quasimodal
hazudesu takes sono, which is a pronominal of kesa tyoosyokuo tabeta
‘eating breakfast this morning’, as a complement. The data clearly
shows that the quasimodals in Japanese are complementtaking

predicates, and in our analysis, soo corresponds to the XCOMP
introduced by a lexical verb, and a quasimodal takes it as a
complement.
Conclusion
On the surface, modality is morphosyntactically encoded in various
ways across languages. However, the conceptual consideration and
empirical observation presented in this paper have revealed that the
modal expressions can be divided into two types, as reflected in the
generative literature. Since LFG postulates parallel syntactic
structures, it enables us to capture different morphosyntactic
behaviours exhibited by different types of modal expressions at two
distinct syntactic levels, cstructure and fstructure. As argued in
Falk , English has cstructurally similar, but fstructurally
different, auxiliaries. I have also illustrated that the two distinct types
of modal expressions in Japanese require two different analyses in
LFG. The analyses correctly predict the behaviours observed in the
use of a verbal proform and a pronominal in relation to the quasi
modals. The present study suggests that the microvariants of phrase
structures and functions of modal expressions found in a diverse
range of languages may be captured in a parallel architecture posited
in LFG, which potentially sheds light on typological issues of
modality.
Morphosyntax of Modals and Quasimodals ++0
Notes
I would like to thank Yasunari Harada, Tomoko Ohkuma, Sachiko
Shudo and Hiroshi Umemoto for valuable discussion of the ideas
presented in this paper. This paper is a part of the outcome of research

++/
Morphosyntax of Modals and Quasimodals
performed under a Waseda University Grant for Special Research Projects
Project number : B.
The following abbreviations are used throughout this paper : / /,
first/second/third person ; ADV, adversative ; AUX, auxiliary ; COND,
conditional ; CONJECT, conjective ; COP, copula ; DAT, dative ; DS, different
subject ; FUT, future ; GEN, genitive ; IMP, imperative ; INDIC, indicative ;
INTENT, intentional ; EXCL, exclamative, EXPL, exemplative ; IRR, irrealis ; NEG,
negative ; NPST, nonpast ; OBLIG, obligative ; PASS, passive ; PL, plural ; POL,
polite ; PRES, present ; Q, question ; REPORT, reportative ; SG, singular ; SIM,
simultaneous ; SUBJUNCT, subjunctive.
Quasimodal is not a special term in Japanese linguistics. Chapin ,
for example, uses this term to refer to a set of periphrastic modal
expressions in English see also Collins .
The cstructure configuration of Japanese is rather controversial.
Although most of the works on Japanese in LFG literature generally agree
that Japanese lacks functional projections like IP, whether it has a VP or
not is an open question. In this paper, I adopt the VP structure, but
assuming the nonVP structure would not affect the argument.
References
Bresnan, J. . Lexicalfunctional syntax. Oxford : Blackwell.
Bybee, J., Perkins, R. & Pagliuca, W. . The evolution of grammar : Tense,
aspect, and modality in the languages of the world. Chicago : The University of
Chicago Press.
Chapin, P.G. . Quasimodals. Journal of Linguistics . .
Chomsky, N. . Aspects of the theory of syntax. Cambridge, MA : MIT Press.
Cinque, G. . Adverbs and functional heads : A crosslinguistic perspective.
Oxford : Oxford University Press.
Collins, P. . Modals and quasimodals in English. Amsterdam : Rodopi.
Dyvik, H. . The universality of fstructure : Discovery or stipulation?
the case of modals. In Butt, M. & King, T.H. eds., Proceedings of the LFG
+333 conference, Stanford, CA : CSLI Publications.
Falk, Y. . Functional relations in the English auxiliary system.
Linguistics . .
Harbert, W. . The Germanic languages. Cambridge : Cambridge

Morphosyntax of Modals and Quasimodals ++.
University Press.
Inoue, K. . Nihongo no modal saiko Japanese modals revisited. In
Hasegawa, N. ed., Nihongo no shubun genshoo Clausal phenomena in
Japanese, Tokyo : Hituzi Shoboo. .
Jackendoff, R. . Toward an explanatory semantic representation.
Linguistic Inquiry . .
Jackendoff, R. . X…syntax : A study of phrase structure. Cambridge, MA :
MIT Press.
Kiefer, F. . Ondefining modality. Folia Linguistica . .
Kroeger, P.R. . Analyzing grammar : An introduction. Cambridge :
Cambridge University Press.
Lyons, J. . Introduction to theoretical linguistics. Cambridge : Cambridge
University Press.
Lyons, J. . Semantics. Cambridge : Cambridge University Press.
Mithun, M. . The languages of Native North America. Cambridge :
Cambridge University Press.
Narrog, H. . Ondefining modality again. Language Sciences . .
Nitta, Y. . Nihongo no modality to ninshoo Japanese modality and person.
Tokyo : Hituzi Shoboo.
Otoguro, R. . Paradigm gaps and periphrases in the Japanese
conjugation system. In Proceedings of western conference on linguistics ,**1.
University of California, San Diego. .
Otoguro, R. . Surfaceoriented morphological network. Humanitas .
. The Waseda University Law Association.
Palmer, F.R. . Mood and modality. Cambridge : Cambridge University
Press, second edition.
Pollock, Y. . Verb movement, Universal Grammar, and the structure of
IP. Linguistic Inquiry . .
Ross, J.R. . Auxiliaries are main verbs. In Todd, W. ed., Studies in
philosophical linguistics, Evanston, IL : Great Expectations Press. .
Schwarze, C. . The syntax of Romance auxiliaries. In Butt, M. & King, T.
H. eds., Proceedings of the LFG30 conference, Stanford, CA : CSLI
Publications.
Ueda, Y. . Nihongo no modality no toogo koozoo to ninsyoo hyoogen
Syntax of Japanese modality and person restrictions. In Hasegawa, N.

Morphosyntax of Modals and Quasimodals ++-
ed., Nihongo no shubun genshoo Clausal phenomena in Japanese, Tokyo :
Hituzi Shoboo. .
Wright, S. . Subjectivity and experiential syntax. In Stein, D. & Wright,
S. eds., Subjectivity and subjectivisation : Linguistic perspectives, Cambridge :
Cambridge University Press. .